CN102384842A - System and method for monitoring health of airfoils - Google Patents

Abstract

The inventon relates to a system and method for monitoring health of airfoils. The method for monitoring the health of a plurality of blades 12 is presented. The method includes the steps of determining 106 delta TOAs corresponding to the plurality of blades 12, determining 404 standard deviation utilizing the delta TOAs corresponding to the plurality of blades; determining 412 delta sigma_1 utilizing the standard deviation and an initial standard deviation, determining 414 normalized delta TOA corresponding to one or more of the plurality of blades utilizing the delta sigma_1, determining 416 standard deviation of the normalized delta TOA, determining 418 delta sigma_2 utilizing the standard deviation of the normalized delta TOA and a previous standard deviation of normalized delta TOA, and determining 418 corrected delta TOA corresponding to the one or more of the plurality of blades based upon the delta sigma_2.

Description

Be used to monitor the system and method for the health status of aerofoil profile part

Technical field

Embodiment of the present disclosure relates generally to the system and method for the health status that is used to monitor spinner blade or aerofoil profile part.

Background technology

Spinner blade or aerofoil profile part are played the part of important role in many devices, several instances comprise axial compressor, turbine, engine, turbine or the like.For example, axial compressor has a series of level, and each level comprises a row's spinner blade or an aerofoil profile part, is a row's stator blade or a quiet aerofoil profile part subsequently.Therefore, each level comprises a pair of spinner blade or aerofoil profile part and quiet aerofoil profile part.Typically, spinner blade or aerofoil profile part can improve the kinetic energy that gets into the fluid of axial compressor through inlet.In addition, stator blade or quiet aerofoil profile part convert the kinetic energy of the raising of fluid to static pressure through diffusion substantially.Therefore, spinner blade or aerofoil profile part and quiet aerofoil profile part are played the part of important role for the pressure that improves fluid.

In addition and since comprise the aerofoil profile part axial compressor widely with application miscellaneous, spinner blade or aerofoil profile part and quiet aerofoil profile part are more important.Axial compressor for example can be used in many devices, for example, and based on the gas turbine on ground, jet engine, high speed marine engine, Compact Power Plant or the like.In addition, axial compressor can be used in the application miscellaneous, for example, and high capacity air separation plant, blast furnace air, fluid catalytic cracking air, dehydrogenating propane or the like.

The aerofoil profile part influences extreme the moving than long hourage down with service condition miscellaneous (high pressure and high temperature for example at a high speed) of the health status of aerofoil profile part in meeting.With the condition miscellaneous, some other factors can cause the fatigue and the stress of aerofoil profile part except extreme.These factors for example can comprise resonant frequency, the vibration in the aerofoil profile part, vibration stress, the temperature stress of inertial force (comprising centrifugal force), pressure, aerofoil profile part, load that moves seat (reseating), gas or other fluid of aerofoil profile part or the like.Stress causes defective and crack with tired long-term increase meeting in the certain hour section in the aerofoil profile part.One or more can broadening along with the time in the crack causes the release of the part of aerofoil profile part or aerofoil profile part.The release of aerofoil profile part can be dangerous for the device that comprises the aerofoil profile part, and therefore can cause huge monetary loss.In addition, it is dangerous and fearful near the personnel device.

Therefore, it is very desirable developing the system and method that can predict the health status of aerofoil profile part in real time.More specifically, develop and to predict in real time that crack or the system and method that breaks are desirable.

Summary of the invention

In brief, according to the one side of present technique, a kind of method that is used to monitor the health status of a plurality of blades is provided.This method comprises the Δ TOA that confirms corresponding to these a plurality of blades; Use is corresponding to the Δ TOA of these a plurality of blades deviation that settles the standard; Use this standard deviation and primary standard deviation to confirm Δ σ _ 1; Use Δ σ _ 1 to confirm one or more normalization Δ TOA corresponding to these a plurality of blades; Confirm the standard deviation of normalization Δ TOA, use the standard deviation of normalization Δ TOA and the standard deviation before the normalization Δ TOA to confirm Δ σ _ 2, and confirm this one or more correction Δ TOA corresponding to these a plurality of blades based on Δ σ _ 2.

According to present technique on the other hand, a kind of method that is used to monitor the health status of a plurality of blades is provided.This method may further comprise the steps: confirm the Δ TOA corresponding to these a plurality of blades; Calculating is corresponding to the standard deviation of the Δ TOA of these a plurality of blades; Confirm one or more normalization Δ TOA corresponding to these a plurality of blades; Confirm the standard deviation of normalization Δ TOA, use the standard deviation of normalization Δ TOA and the standard deviation before the normalization Δ TOA to confirm Δ σ _ 3, and confirm to revise Δ TOA based on Δ σ _ 2.

According on the one hand, a kind of system that comprises processing subsystem is provided.This processing subsystem is confirmed the Δ TOA corresponding to a plurality of blades; Use is corresponding to the Δ TOA of these a plurality of blades deviation that settles the standard; Use standard deviation and primary standard deviation to confirm Δ σ _ 1; Use this Δ σ _ 1 to confirm one or more normalization Δ TOA corresponding to these a plurality of blades; Confirm the standard deviation of normalization Δ TOA, use the standard deviation of normalization Δ TOA and the standard deviation before the normalization Δ TOA to confirm Δ σ _ 2, and confirm one or more correction Δ TOA corresponding to these a plurality of blades based on Δ σ _ 2.

According on the other hand, a kind of system that comprises processing subsystem is provided.This disposal system is confirmed the Δ TOA corresponding to a plurality of blades; Calculating is corresponding to the standard deviation of the Δ TOA of these a plurality of blades; Confirm one or more normalization Δ TOA corresponding to these a plurality of blades; Confirm the standard deviation of normalization Δ TOA, utilize the standard deviation of normalization Δ TOA and the standard deviation before the normalization Δ TOA to confirm Δ σ _ 3, and confirm to revise Δ TOA based on Δ σ _ 3.

Description of drawings

When describing in detail below with reference to advantages, of the present invention these with understandings that will improve of further feature, aspect and advantage, in the accompanying drawings, similar label parts like the representation class in all are schemed, wherein:

Fig. 1 is the illustrative diagram according to the blade health monitoring system of an embodiment of native system;

Fig. 2 is the process flow diagram of expression according to the static deflection that is used for definite blade with the illustrative methods of dynamic deflection of an embodiment of present technique;

Fig. 3 is the process flow diagram of expression according to the illustrative methods of the static deflection that is used for definite blade of an embodiment of present technique;

Fig. 4 is the process flow diagram of expression according to the illustrative methods of the static deflection that is used for definite blade of another embodiment of present technique;

Fig. 5 is the process flow diagram of expression according to the illustrative methods of the static deflection that is used for definite blade of another embodiment of present technique; And

What Fig. 6 was expression according to an embodiment of present technique is used for confirming the process flow diagram corresponding to the step of the method for moving a side-play amount of blade.

Embodiment

As following detailed argumentation, the embodiment of native system and technology can assess the health status of one or more blades or aerofoil profile part.More specifically, native system and the technology blade confirming to cause or the static deflection of aerofoil profile part owing to the one or more defectives in blade or aerofoil profile part or crack.Hereinafter, will use term " aerofoil profile part " and " blade " convertibly.Original or the desired locations that static deflection for example can be used to refer to blade is with respect to the expectation of blade or the steady change in original position.Some embodiment of native system and technology has also confirmed the dynamic deflection corresponding to blade.As used herein, term " dynamically deflection " can be used to refer to the amplitude of the vibration of the blade on the mean place of blade.

Be in operation, because one or more cracks or defective in blade, blade can be different from expectation TOA in the time of arrival (TOA) at place, reference position.Therefore, the variation of the TOA of blade can be used to confirm the static deflection of blade.As used herein, term " expectation TOA " can be used to refer to that defective or crack and blade are not worked in the ideal case when in blade, having, load condition for the best and vibration in blade hour blade at the TOA at place, reference position.Hereinafter, for easy understanding, will use word " TOA " and term " actual TOA " convertibly.

But except crack or defective in blade, TOA also can change owing to the seat that moves of one or more service datas and blade.Service data for example can comprise inlet guide vane (IGV) angle, load, speed, mass rate, discharge pressure or the like.As used herein, term " blade move seat " can be used to refer to blade and is locked in the joint (for example dovetail joint) in the position of the original or desired locations that is different from blade.Typically, blade pass is crossed one or more joints (for example dovetail joint) and is fastened on the rotor.Between the starting period of the device that comprises blade, blade can offset from they original positions joint, and can be locked in the joint in the position in the original position that is different from blade.As an example, this device can comprise gas turbine, compressor or the like.Blade is locked in the seat that moves that is called as blade in the joint in the position in the original position that is different from blade.The variation of the position of blade can change the actual TOA of blade.

Therefore, because the influence that moves seat of service data and blade, based on the actual TOA of blade and definite static deflection can be because crack in blade or defective former thereby change or above definite or accurate static deflection.Therefore, cancellation service data and blade move seat to the influence of actual TOA to confirm that definite static deflection (being " static deflection " hereinafter) is very important.Some embodiment of present technique has cancelled the influence that moves seat of the one or more and blade in the service data and has confirmed static deflection from the actual TOA of blade.Other embodiment of some of present technique can make service data that the influence of actual TOA is standardized or compensate this influence.

Fig. 1 is the synoptic diagram according to the rotor blade health monitoring system 10 of an embodiment of native system.As shown in fig. 1, system 10 comprises one or more blades or aerofoil profile part 12, and they are by system's 10 monitorings, to confirm the static deflection of blade 12.In certain embodiments, system 10 also confirms the dynamic deflection corresponding to blade 12.Shown in the current structure of conceiving, system 10 comprises one or more sensors 14,16.In the sensor 14,16 each produces the TOA signal 18,20 of the actual TOA that is illustrated on the special time period at the blade 12 of datum respectively.In one embodiment, the one or more blades 12 of sensor 14,16 sensings arrive datum and produce TOA signal 18,20.Reference point for example can be at sensor 14, below 16 or proximity sense 14,16.In one embodiment, each in the TOA signal 18,20 sampled and/or measured reaches special time period, and is used to confirm the actual TOA of blade.For example can measure actual TOA by the unit of the time or the number of degrees.

In one embodiment, but the arrival of the leading edge of the one or more blades 12 of sensor 14,16 sensings and produce TOA signal 18,20.In another embodiment, but the arrival of the trailing edge of the one or more blades 12 of sensor 14,16 sensings and produce signal 18,20.In yet another embodiment, but the arrival of the leading edge of the one or more blades 12 of sensor 14 sensings and produce TOA signal 18, but and the arrival of the trailing edge of the one or more blades 12 of sensor 16 sensings and produce TOA signal 20, or vice versa. Sensor 14,16 for example can be close to one or more blades 12 and be installed on the fixed object in a kind of like this position: this position makes the arrival of the one or more blades 12 of sensing efficiently.In one embodiment, at least one sensor 14,16 is installed on the housing (not shown) of one or more blades 12.As limiting examples, sensor 14,16 can be magnetic sensor, capacitive transducer, eddy current sensor or the like.

As shown in the structure of current design, TOA signal 18,20 is received by processing subsystem 22.Processing subsystem 22 is confirmed the actual TOA of one or more blades 12 based on TOA signal 18,20.In addition, processing subsystem 22 is confirmed the static deflection of these one or more blades 12 based on the actual TOA of one or more blades 12.More specifically, processing subsystem 22 is configured to so that confirm the one or more static deflection in the blade 12 through the actual TOA that handles one or more blades 12.As noted earlier, the actual TOA of blade 12 can be influenced by the seat that moves of one or more service datas and blade 12.

Therefore, because service data is to the seat that moves of the influence of actual TOA and blade 12, the static deflection definite based on the actual TOA of one or more blades 12 can be exaggerative value.For example; Since service data and blade 12 move the influence of seat to the actual TOA of blade 12, the static deflection definite based on the actual TOA of blade 12 can be presented at one or more defectives or crack (even if when in blade 12, not having crack or defective) in the blade 12.Therefore, in one embodiment, processing subsystem 22 can be confirmed the influence of one or more service datas to the actual TOA of one or more blades 12.In addition, processing subsystem 22 is confirmed static deflection through deducting one or more service datas to the influence of the actual TOA of one or more blades 12.As noted earlier, service data can comprise variation that moves seat, asynchronous vibration, synchronous vibration, speed, temperature, speed of inlet guide vane (IGV) angle, load variations, blade or the like.Processing subsystem 22 for example can receive the service data from field monitoring machine (OSM) 24, and OSM monitors service data through sensor, video camera and other device.In addition, processing subsystem 22 makes the seat that moves of blade standardize to the influence of the actual TOA of blade 12.To come to set forth in more detail through the deduction service data to the influence of actual TOA or make this influence normalization confirm static deflection with reference to Fig. 2-5.Processing subsystem 22 also is configured to so that confirm the dynamic deflection corresponding to one or more blades 12 based on the static deflection and the actual TOA of one or more blades 12.In one embodiment, processing subsystem 22 can have data warehouse 26, data warehouse 26 storage data, for example, static deflection, dynamically deflection, TOA, Δ TOA, any intermediate data or the like.

Referring now to Fig. 2, described to represent the process flow diagram that is used for static deflection with the illustrative methods 100 of dynamic deflection of definite one or more blades according to an embodiment of the invention.These one or more blades for example can be one or more blade 12 (see figure 1)s.This method begins at step 102 place, wherein, can receive by processing subsystem corresponding to each the TOA signal in one or more blades, for example, processing subsystem 22 (see figure 1)s.As the front was mentioned with reference to Fig. 1, the TOA signal can be produced by sensor, for example sensor 14,16 (see figure 1)s.In addition, the TOA signal for example can be TOA signal 18,20.

In addition, at step 104 place, confirm by processing subsystem corresponding to each the actual TOA in one or more blades.The processing subsystem utilization is confirmed actual TOA corresponding to each the TOA signal in one or more blades.More specifically, the processing subsystem utilization is confirmed the one or more actual TOA corresponding to blade corresponding to the TOA signal of blade.At step 106 place, can confirm corresponding to each the Δ TOA in one or more blades.Corresponding to the Δ TOA of blade for example can be confirm at step 104 place corresponding to the actual TOA of blade and poor corresponding to the expectation TOA105 of blade.It may be noted that Δ TOA corresponding to blade is illustrated in the variation with respect to the expectation TOA105 of blade at certain hour constant place.For example can use following equality (1) to confirm Δ TOA:

ΔTOA _k(t)＝TOA _act(k)(t)-TOA _exp(k) (1)

Wherein, Δ TOA _k(t) be time constant t place corresponding to the Δ TOA of blade k or time constant t place with respect to the variation corresponding to the expectation TOA of blade k, TOA _{Act (k)}Be actual TOA corresponding to blade k at time constant t place, and TOA _{Exp (k)}Be expectation TOA corresponding to blade k.

As used herein, term " expectation TOA " can be used to refer to the actual TOA when the blade that in blade, does not have defective or crack and blade when service data is worked in to the influence of actual TOA running status hour, to locate in the reference position.In one embodiment, when nearest trust manufacturing or when having bought the device that comprises blade, can confirm expectation TOA through the expectation TOA that makes actual TOA corresponding to blade equal blade corresponding to blade.This definite hypothesis is because entrust manufacturing or bought device recently, so all blades are all worked in the ideal case, load condition is best, and the vibration in blade is minimum.In another embodiment, can be through being averaged confirm expectation TOA the actual time of arrival (TOA) of all blades in device.Device for example can comprise axial compressor, the gas turbine based on ground, jet engine, high speed marine engine, Compact Power Plant or the like.It may be noted that the unit representation of Δ TOA with the time or the number of degrees.

In one embodiment, at step 108 place, be convertible into the unit of Mill corresponding to each the unit of Δ TOA in one or more blades.In one embodiment, can use following equality (2) will corresponding in one or more blades each convert the unit of Mill to as the Δ TOA of unit with the number of degrees:

$\mathrm{\Δ}{\mathrm{ToA}}_{\mathrm{mils}\left(k\right)}\left(t\right)=\frac{2\mathrm{\πR}\×\mathrm{\Δ}{\mathrm{ToA}}_{\mathrm{Deg}\left(k\right)}\left(t\right)}{360}---\left(2\right)$

Wherein, Δ TOA _{Mils (k)}(t) be Δ TOA, and Δ TOA is unit with the Mill, Δ ToA at the blade k at time constant t place _{Deg (k)}(t) be Δ TOA, and Δ TOA is unit with the number of degrees, and R is the radius that the tip from centre of rotor to blade k is measured at the blade k at time constant t place.Radius R is unit with the Mill.In another embodiment, can use following equality (3) will convert the unit of Mill to as the Δ TOA of unit with second:

$\mathrm{\Δ}{\mathrm{ToA}}_{\mathrm{mils}\left(k\right)}\left(t\right)=\frac{2\mathrm{\πR}\×N\×\mathrm{\ΔT}{\mathrm{oA}}_{\sec \left(k\right)}\left(t\right)}{60}---\left(3\right)$

Wherein, Δ ToA _{Mils (k)}(t) be Δ TOA, and Δ TOA is unit with the Mill, Δ ToA at the blade k at time constant t place _{Sec (k)}(t) be Δ TOA, and Δ TOA is unit with the number of degrees, and R is the radius from the centre of rotor of this blade of blade at the blade k at time constant t place.Radius R is unit with the Mill.

In addition, at step 110 place, confirm each the static deflection in one or more blades based on Δ TOA.To set forth the confirming of static deflection of one or more blades with reference to Fig. 3-5 in more detail.At step 112 place, can confirm dynamic deflection subsequently corresponding to one or more blades.In one embodiment, can confirm dynamic deflection through from Δ TOA, deducting corresponding to the static deflection of blade corresponding to blade corresponding to blade.In another embodiment, can confirm dynamic deflection through deducting corresponding to the static deflection of blade the Δ TOA that filters from warp corresponding to blade corresponding to blade.For example can be through the Δ TOA corresponding to blade that confirms at step 106 place being filtered to confirm Δ TOA through filtering.The one or more technology of average filtration, median filter or the like that comprise capable of using come Δ TOA is filtered.

As noted earlier, the actual TOA of one or more blades can be used to confirm the static deflection of blade.But, the moving seat and can influence the actual TOA of blade of one or more service datas and blade.Therefore, the static deflection of confirming based on the actual TOA of blade possibly not be a static deflection accurately.Therefore, for confirming definite static deflection, remove or deduct one or more service datas and blade to move an influence to actual TOA be necessary.To set forth the illustrative methods that is used for confirming static deflection with reference to Fig. 3 through an influence that moves of the one or more service datas of Δ TOA deduction confirmed from actual TOA or based on this actual TOA and blade.Referring now to Fig. 3, described to represent the process flow diagram of the illustrative methods 110 of the static deflection that is used for confirming blade according to an embodiment of the invention.The step 110 of Fig. 2 more specifically, has been described according to the illustrative aspects of present technique in more detail.

As shown in Figure 3, reference number 302 expressions are corresponding to the Δ TOA of blade.In one embodiment, the described technology of the step 106 with reference to Fig. 2 capable of using is confirmed Δ TOA302.In addition, at step 304 place, can receive corresponding to blade or comprise one or more service datas of the device of blade.As noted earlier, service data for example can comprise (IGV) angle, load, temperature, speed, mass rate, discharge pressure or the like.Service data for example can be received from field monitoring device 24 (see figure 1)s by processing subsystem 22.

In addition, at step 306 place, can carry out and check whether blade is operation for the first time after the startup of the device that comprises blade.At step 306 place, be operation for the first time after startup if confirm blade, then control can be transferred to step 308.At step 308 place, confirm one or more coefficients based on one or more parts of service data.Following equality for example capable of using (4) is confirmed coefficient:

$\mathrm{\Δ}{\mathrm{TOA}}_k=\stackrel{\‾}{A}\stackrel{\‾}{D}---\left(4\right)$

Wherein, Δ TOA _kBe the Δ TOA of blade k,

Be one or more parts of service data, and

It is coefficient.Coefficient is confirmed in the linear combination of one or more parts that in one embodiment, can be through forming service data.In addition, the value of one or more parts of service data can be replaced, to confirm coefficient.In addition, at step 312 place, the coefficient of having confirmed at step 308 place is stored in the data warehouse, for example data warehouse 26 (see figure 1)s.It may be noted that when coefficient storage is in data warehouse any other the existing coefficient in data warehouse can be eliminated.

Referring back to step 306, is not operation for the first time after startup if confirm blade, and then control can be transferred to step 310.At step 310 place, from data warehouse, fetch coefficient.Fetching coefficient at step 310 place is to have supposed to comprise to have confirmed between starting period of device of blade that coefficient and coefficient are therefore Already in the data warehouse.At step 314 place, can confirm the influence that causes owing to the IGV angle subsequently to Δ TOA302.In one embodiment, can use following exemplary equation (5) to confirm the influence that causes owing to IGV:

T _IGV(t)＝f(IGV(t)) (5)

Wherein, T _IGV(t) be that IGV (t) is at the IGV angle at time constant t place in the IGV at the time constant t place influence to Δ TOA, and f is the function of IGV (t).In one embodiment, can be through confirming IGV (t) and confirming the function of IGV corresponding to the multiplier of the coefficient of IGV (t).

At step 316 place, can confirm the influence that causes owing to load to Δ TOA302.Following equality capable of using (6) comes definite owing to the influence to Δ TOA302 of loading and causing:

T _load(t)＝g(DWATT(t)) (6)

Wherein, T _Load(t) be that DWATT is the load at time constant t place in the influence to Δ TOA of the load at time constant t place, and g is the function of load.In one embodiment, can be through confirming DWATT (t) and confirming the function of DWATT corresponding to the multiplier of the coefficient of DWATT.In another embodiment, multiplier that can be through confirming DWATT (t) and coefficient and confirm the function of DWATT corresponding to the linear combination of another coefficient of DWATT.

Subsequently, at step 318 place, can confirm the influence that causes owing to temperature in (CTIM) to Δ TOA302.Following equality capable of using (7) come to be confirmed the influence that causes owing to temperature in (CTIM):

T _CTIM(t)＝d(CTIM(t)) (7)

Wherein, T _CTIMBe the value to the influence of Δ TOA that causes owing to temperature at time constant t place, CTIM (t) is the temperature at t thermal constant place, and d is the coefficient corresponding to temperature in.When the influence of confirming at step 314 place to cause to Δ TOA302 owing to IGV, confirm load to the influence of Δ TOA302 and after step 318 place confirms the influence of CTIM to Δ TOA302 at step 316 place, confirm normalization Δ TOA at step 320 place.For example can confirm normalization Δ TOA through from Δ TOA302, deducting such as the influence of the service data of IGV, load and temperature in (CTIM).

In one embodiment, for example can use following exemplary equation (8) to confirm normalization Δ TOA:

Norm_ΔTOA _k(t)＝ΔTOA _k(t)-T _load(t)-T _CTIM(t)-T _IGV(t) (8)

Wherein, Norm_ Δ TOA _k(t) be normalization Δ TOA, Δ TOA corresponding to blade k at time constant t place _k(t) be Δ TOA corresponding to blade k at time constant t place, and T _Load(t), T _CTIM(t), T _IGV(t) be respectively in load, temperature in and the IGV at time constant t place influence to Δ TOA.

Typically, one or more blade pass are crossed one or more joints (for example dovetail joint) and are fastened on the rotor.Between the starting period of the device that comprises blade, blade can offset from they original positions joint, and can be locked in the joint in the position in the original position that is different from blade.Blade is locked in the seat that moves that is called as blade in the joint in the position in the original position that is different from blade.The variation of the position of blade can change the actual TOA of blade.Therefore, actual TOA the Δ TOA that confirms and the Δ TOA that standardizes based on blade possibly be inaccurate.More specifically, owing to the seat that moves of blade, Δ TOA and normalization Δ TOA possibly be inaccurate.Therefore, revise actual TOA, Δ TOA or normalization Δ TOA corresponding to blade to remove since blade to move an influence that causes be necessary.Step 322-330 has revised the normalization Δ TOA definite at step 320 place and the Δ TOA302 of blade, to remove the influence that seat causes that moves owing to blade.

At step 322 place, can carry out and check whether blade is operation for the first time after startup.At step 322 place, be operation for the first time after startup if confirm blade, then control can be transferred to step 324.At step 324 place, can confirm a side-play amount of moving corresponding to blade.Own like this paper, term " moves a side-play amount " and can be used to refer to a numerical value of the influence that causes that moves of removing the Δ TOA that can be used for from blade, actual TOA or the normalization Δ TOA owing to blade.To set forth in more detail with reference to Fig. 6 and move confirming of a side-play amount.Subsequently, at step 326 place, that confirms at step 324 place moves a side-play amount and can be stored in the data warehouse.Moving a side-play amount for example can be stored in data warehouse 26 (see figure 1)s.Can notice; In the structure of current design; When blade moves after startup for the first time, confirm to move a side-play amount, may be locked in such position because supposed blade: this position is different from the original position of this blade between the starting period of the device that comprises this blade.

Referring back to step 322, is not operation for the first time after the startup of the device that comprises this blade if confirmed blade, and then control can be transferred to step 328.It may be noted that when blade be not when after startup, moving for the first time, confirmed a side-play amount of moving after being illustrated in the startup of the device that comprises blade, and this moves a side-play amount and is stored in the data warehouse corresponding to this blade.Therefore, at step 328 place, can from data warehouse, fetch a side-play amount of moving corresponding to blade.

When move in step 326 place storage a side-play amount or step 328 place fetch move a side-play amount after, can confirm correction Δ TOA at step 330 place.In one embodiment, can confirm to revise Δ TOA through a normalization Δ TOA that correction has been confirmed at step 320 place that moves to blade.For example can move a side-play amount and confirm to revise Δ TOA through from normalization Δ TOA, deducting corresponding to blade.In another embodiment, can confirm to revise Δ TOA through revising Δ TOA302.In this embodiment, can move a side-play amount and confirm to revise Δ TOA through from Δ TOA302, deducting corresponding to blade.In addition, at step 332 place, can filter revising Δ TOA, to produce static deflection 334.The filtration of revising Δ TOA can reduce the noise from this correction Δ TOA.For example can use median filter, moving average filtration or their combination to come to filter to revising Δ TOA.

As above-mentioned, one or more service datas can influence the actual TOA of a plurality of blades.But service data possibly can't influence the actual TOA of blade equably.Therefore, compare with the actual TOA of other blade in a plurality of blades, the one or more actual TOA in the blade can be affected more.Therefore, compare, because the additional influence of service data can show defective or crack in the blade corresponding to the one or more static deflection in the blade with static deflection corresponding to other blade.In addition, the static deflection of confirming based on the actual TOA of blade possibly not be a static deflection accurately.Therefore, making service data is necessary to the influence normalization of the actual TOA of a plurality of blades in device.To set forth with reference to Figure 4 and 5 and be used for through making one or more service datas standardize to confirm the illustrative methods of static deflection to actual TOA or based on the influence of the definite Δ TOA of this actual TOA.

Referring now to Fig. 4, described expression according to the illustrative methods 110 that is used for confirming static deflection of another embodiment ' the process flow diagram of step.More specifically, Fig. 4 has set forth the step 110 of Fig. 2 according to an embodiment of the present technique that is used for definite static deflection.As shown in Figure 4, reference number 402 expression is corresponding in the Δ time of arrival (TOA) such as a plurality of blades in the device of turbine, axial compressor etc.The technology that step 106 with reference to Fig. 2 capable of using is set forth is confirmed corresponding to each the Δ TOA in these a plurality of blades.In one embodiment, Δ TOA402 can be similar to the Δ TOA that confirms at step 106 place of Fig. 2.

In addition, at step 404 place, can calculate standard deviation corresponding to the Δ TOA of a plurality of blades.For example, comprise in five blades and five blades each having Δ TOA-it is Δ TOA when these a plurality of blades ₁, Δ TOA ₂, Δ TOA ₃, Δ TOA ₄, Δ TOA ₅, then can calculate Δ TOA at step 404 place ₁, Δ TOA ₂, Δ TOA ₃, Δ TOA ₄With Δ TOA ₅ Standard deviation.At step 406 place, can carry out inspection and confirm whether blade is operation for the first time after the startup of the device that comprises these a plurality of blades subsequently.At step 406 place, be operation for the first time after startup if confirm blade, then control can be transferred to step 408.

For easy understanding, term " standard deviation " will be called " current standard deviation " hereinafter.As shown in fig. 4, at step 408 place, the standard deviation that calculates at step 404 place can be used as primary standard deviation 410 and stores.Primary standard deviation 410 can be stored in the data warehouse 26 of data warehouse-for example.As used herein, term " primary standard deviation " can be used as the current standard deviation of when blade brings into operation for the first time, confirming and quotes after startup.More specifically, the standard deviation of confirming at step 404 place can be used as primary standard deviation 410 and is stored in the data warehouse.

Referring back to step 406, is not operation for the first time after startup if confirm blade, and then control can be transferred to step 412.At step 412 place, current standard deviation and the primary standard deviation of confirming at step 404 place 410 capable of using confirmed Δ σ _ 1.More specifically, can confirm Δ σ _ 1 through confirming current standard deviation and the difference between the primary standard deviation 410 confirmed at step 404 place.It may be noted that when after the startup of the device that is comprising a plurality of blades for the first time during treatment step 412 then the value of primary standard deviation 410 and the current standard deviation confirmed at step 404 place equates.Therefore, at step 412 place, the value of Δ σ _ 1 can equal zero.

In addition, at step 414 place, can confirm corresponding to the one or more normalization Δ TOA in a plurality of blades.For example can confirm normalization Δ TOA based on following equality (9):

Norm_ΔTOA _k(t)＝ΔTOA _k(t)-K*(Δσ(t)_1)-Mean(ΔTOA _1toj(t)) (9)

Wherein, Norm_ Δ TOA _k(t) be normalization Δ TOA, Δ TOA corresponding to blade k at time constant t place _k(t) be Δ TOA corresponding to blade k at time constant t place, and Δ σ (t) _ the 1st, Δ σ _ 1 at time constant t place, and K is a constant.In one embodiment, can be based on the value of confirming constant K corresponding to the average of the Δ TOA of blade.In one embodiment, the value of K can be 1.In another embodiment, the value of K can be-1.In yet another embodiment, the value of K can be 0.

In addition, at step 416 place, can confirm current standard deviation corresponding to the one or more normalization Δ TOA in a plurality of blades.At step 418 place, can confirm Δ σ _ 2 subsequently.For example can confirm Δ σ _ 2 through confirming the difference between the standard deviation before the current standard deviation of normalization Δ TOA and the Δ TOA that standardizes.Term " standard deviation before of normalization Δ TOA " can be used to refer to the current standard deviation of comparing the normalization Δ TOA that confirms at time step T-1 place with the current standard deviation of the normalization Δ TOA that confirms at time step T place.

After definite Δ σ _ 2,, can carry out and check whether whether Δ σ _ 2 are operations for the first time after startup greater than predetermined first threshold and/or a plurality of blades at step 420 place.Can confirm predetermined first threshold with the mode of experience based on historical Δ TOA corresponding to blade.At step 420 place, are operations for the first time after startup if confirm Δ σ _ 2 greater than predetermined first threshold or this a plurality of blades, then control can be transferred to step 422.At step 422 place, can confirm corresponding in these a plurality of blades this one or more move a side-play amount.To come to set forth in more detail to move confirming of a side-play amount with reference to Fig. 6.After confirming to move a side-play amount,, move a side-play amount and can be stored in data warehouse 26 (see figure 1)s of data warehouse-for example at step 424 place.

Referring back to step 420, is not when after startup, moving for the first time when definite Δ σ _ 2 are not more than predetermined first threshold and a plurality of blade, and then control can be transferred to step 426.At step 426 place, can from data warehouse, fetch and move a side-play amount.It may be noted that when Δ σ _ 2 are not more than predetermined first threshold and blade be not when after startup, moving for the first time, do not produce and move a side-play amount.Therefore, at step 426 place, fetch an existing side-play amount of moving from data warehouse.Fetch move a side-play amount after, can confirm at step 428 place corresponding to the one or more correction Δ TOA in a plurality of blades.The technology of setting forth with reference to the step 330 of Fig. 3 for example capable of using confirms to revise Δ TOA.As the front was mentioned with reference to Fig. 3, the technology of setting forth with reference to the step 330 of Fig. 3 capable of using confirmed to revise Δ TOA.For example, capable of using confirm at step 414 place confirm correction Δ TOA corresponding to the normalization Δ TOA of blade and the side-play amount of moving corresponding to this blade of from data warehouse, fetching corresponding to this blade at step 426 place.In one embodiment, can confirm correction Δ TOA through from Δ TOA, deducting corresponding to a side-play amount of moving of this blade corresponding to blade corresponding to this blade.Δ TOA for example can be corresponding to the Δ TOA402 of these a plurality of blades one of them.

In addition, at step 430 place, can filter revising Δ TOA, to produce corresponding to the one or more static deflection 432 in these a plurality of blades.As the front was mentioned with reference to Fig. 3, the filtration of revising Δ TOA can reduce from the noise of revising Δ TOA.For example can use median filter technology, moving average filtering technique or their combination to come to filter to revising Δ TOA.

Referring now to Fig. 5, described the illustrative methods 110 that be used for confirm static deflection of expression according to another embodiment " the process flow diagram of step.More specifically, Fig. 5 has set forth the step 110 of Fig. 2 according to an embodiment of the present technique that is used for definite static deflection.As shown in fig. 5, reference number 502 expression is corresponding in the Δ time of arrival (TOA) such as a plurality of blades in the device of turbine, axial compressor etc.The technology of setting forth with reference to the step 106 of Fig. 2 capable of using is confirmed corresponding to each the Δ TOA in these a plurality of blades.In one embodiment, Δ TOA502 can be similar to the Δ TOA that confirms at step 106 place of Fig. 2.

In addition, at step 504 place, can calculate standard deviation corresponding to the Δ TOA of a plurality of blades.For example, comprise in five blades and this five blades each having Δ TOA-it is Δ TOA when a plurality of blades ₁, Δ TOA ₂, Δ TOA ₃, Δ TOA ₄, Δ TOA ₅The time, then can confirm Δ TOA at step 504 place ₁, Δ TOA ₂, Δ TOA ₃, Δ TOA ₄With Δ TOA ₅ Standard deviation.At step 506 place, can confirm subsequently corresponding to the one or more normalization Δ TOA in these a plurality of blades.For example can confirm normalization Δ TOA based on following equality (10):

Norm_ΔTOA _k(t)＝(ΔTOA _k(t)-Mean?ΔTOA _1toj(t))/s?tan?dard_deviation(t) (10)

Wherein, Norm_ Δ TOA _k(t) be normalization Δ TOA, Δ TOA corresponding to blade k at time constant t place _k(t) be Δ TOA, Mean Δ TOA corresponding to blade k at time constant t place _1toj(t) be average corresponding to the Δ TOA of blade 1 to j (comprising blade k).

In addition, at step 508 place, can confirm standard deviation corresponding to the one or more normalization Δ TOA in a plurality of blades.At step 510 place, can confirm Δ σ _ 3 subsequently.For example can confirm Δ σ _ 3 through confirming the difference between the standard deviation before the standard deviation of normalization Δ TOA and the Δ TOA that standardizes.Term " standard deviation before of normalization Δ TOA " can be used to refer to the standard deviation of comparing the normalization Δ TOA that confirms at time step T-1 place with the standard deviation of the normalization Δ TOA that confirms at time step T place.

After step 510 place has confirmed Δ σ _ 3, can carry out at step 512 place and check whether whether Δ σ _ 3 be after startup, to move for the first time greater than predetermined second threshold value and/or a plurality of blades.Can confirm predetermined second threshold value with the mode of experience based on historical Δ TOA.At step 512 place, be operation for the first time after startup if confirm Δ σ _ 3 greater than predetermined second threshold value or a plurality of blade, then control can be transferred to step 514.At step 514 place, can confirm corresponding among one or more in a plurality of blades each move a side-play amount.To set forth in more detail with reference to Fig. 6 and move confirming of a side-play amount.After confirming to move a side-play amount,, move a side-play amount and can be stored in data warehouse 26 (see figure 1)s of data warehouse-for example at step 516 place.

Referring back to step 512, is not when after startup, moving for the first time when definite Δ σ _ 3 are not more than predetermined second threshold value and a plurality of blade, and then control can be transferred to step 518.At step 518 place, can from data warehouse, fetch corresponding among one or more in a plurality of blades each move a side-play amount.It may be noted that when Δ σ _ 3 are not more than predetermined second threshold value and blade be not when after startup, moving for the first time, do not produce and move a side-play amount.Therefore, at step 518 place, fetch an existing side-play amount of moving from data warehouse.Fetch move a side-play amount after, can confirm at step 520 place corresponding to the one or more correction Δ TOA in a plurality of blades.The technology of setting forth with reference to the step 330 of Fig. 3 for example capable of using confirms to revise Δ TOA.As the front was mentioned with reference to Fig. 3, the described technology of the step 330 with reference to Fig. 3 capable of using confirmed to revise Δ TOA.For example, capable of using confirm at step 506 place confirm correction Δ TOA corresponding to the normalization Δ TOA of blade and the side-play amount of moving corresponding to blade of from data warehouse, fetching corresponding to blade at step 518 place.In one embodiment, can confirm correction Δ TOA through from normalization Δ TOA, deducting corresponding to a side-play amount of moving of this blade corresponding to blade corresponding to this blade.In another embodiment, can confirm correction Δ TOA through from Δ TOA, deducting corresponding to a side-play amount of moving of this blade corresponding to blade corresponding to this blade.Δ TOA for example can be corresponding to one among the Δ TOA502 of these a plurality of blades.

In addition, in step 522, can filter revising Δ TOA, to produce static deflection 524.As the front was mentioned with reference to Fig. 3, the filtration of revising Δ TOA can reduce from the noise of revising Δ TOA.For example can use median filter technology, moving average filtering technique or their combination to come to filter to revising Δ TOA.

Referring now to Fig. 6, described expression and be used for producing process flow diagram corresponding to the step of the method for moving a side-play amount 600 of blade according to an embodiment of present technique.More specifically, method 600 has been set forth the step 328 of Fig. 3, the step 422 of Fig. 4 and the step 514 of Fig. 5.As shown in figure 6, reference number 602 expressions are corresponding to normalization Δ time of arrival (TOA) of blade.In one embodiment, normalization Δ TOA602 can be one or more among the normalization Δ TOA that use confirms with reference to the described technology of step 506 of the step 414 of the step 320 of Fig. 3, Fig. 4, Fig. 5.In one embodiment, normalization Δ TOA602 is definite one or more corresponding among the normalization Δ TOA of blade the transient affair of blade after.Transient affair for example can comprise: comprise the lasting variation or the like of speed of startup or shutdown, the blade of the device of blade.

In addition, reference number 604 expressions are utilized in the definite one or more correction Δ TOA corresponding to blade of normalization Δ TOA that produce before the transient affair.Transient affair is such transient affair: after this transient affair, confirmed normalization Δ TOA602.At step 606 place, carry out inspection and confirm whether blade is operation for the first time after startup.At step 606 place, be operation for the first time after startup if confirm blade, then step 608 is transferred in control.In addition, at step 608 place, can carry out inspection and confirm whether blade is to move at the basic load place.At step 608 place, not operation if confirm blade at the basic load place, then control can be transferred to step 610.Referring back to step 606, is not operation for the first time after startup if confirm blade, and then control can be transferred to step 610.At step 610 place, shown and moved a side-play amount Already in the data warehouse, data warehouse 26 (see figure 1)s for example corresponding to blade.Therefore, do not confirm to move a side-play amount.

Referring back to step 608, is the operation at the basic load place if confirm blade, and then control can be transferred to step 612.At step 612 place, can confirm first average of one or more normalization Δ TOA602.In addition, at step 614 place, can confirm second average of one or more correction Δ TOA604.After definite first average and second average, can confirm to move a side-play amount 618 through from first average, deducting second average corresponding to blade at step 616 place.

The embodiment of present technique causes confirming the influence of service data to TOA.In addition, present technique has been deducted the influence of service data from TOA, to confirm normalization Δ TOA.In addition, present technique makes the influence normalization of service data to the TOA of blade, to confirm normalization Δ TOA.Normalization Δ TOA can be used for confirming defective or the crack in blade.Some embodiment of present technique also helps the variation of the TOA of the blade that causes that moves that detects owing to blade.And, the health status of confirming to can be used for monitoring blade of normalization Δ TOA.For example, normalization Δ TOA can be used to confirm in blade, whether have one or more cracks.

Will appreciate that, not necessarily all can realize described such target or advantage above all according to any specific embodiment.Therefore; For example; Those skilled in the art will approve, can embody or carry out system described herein and technology in such a way: this mode realizes or optimizes an advantage or one group of advantage teaching herein, and may not realize can be by other target or advantage teaching herein or that recommend.

The embodiment of limited quantity describes the present invention though only combined, and should understand easily, the invention is not restricted to this disclosed embodiment.On the contrary, can modify, not describe before this but any amount of modification, change, the alternative or equivalent arrangements suitable with the spirit and scope of the present invention with combination to the present invention.In addition, though various embodiment of the present invention has been described, should be appreciated that each side of the present invention can comprise among the said embodiment more only.Therefore, the present invention should not regard as and limited by aforementioned description, but only receives the restriction of the scope of accompanying claims.

New and the expectation of being claimed as receives the content of the patent license protection of the U.S. and sees claim.

Claims (10)

1. method that is used to monitor the health status of a plurality of blades (12) comprises:

Confirm (106) Δ TOA corresponding to said a plurality of blades (12);

Utilization is confirmed (404) standard deviation corresponding to the said Δ TOA of said a plurality of blades;

Utilize said standard deviation and primary standard deviation to confirm (412) Δ σ _ 1;

Utilize said Δ σ _ 1 to confirm (414) one or more normalization Δ TOA corresponding to said a plurality of blades;

Confirm the standard deviation of (416) said normalization Δ TOA;

Utilize the said standard deviation of said normalization Δ TOA to confirm (418) Δ σ _ 2 with the standard deviation before the normalization Δ TOA; And

Confirm (418) said one or more correction Δ TOA based on said Δ σ _ 2 corresponding to said a plurality of blades.

2. method according to claim 1 is characterized in that, when said a plurality of blades (12) are to confirm said primary standard deviation when move the first time after startup.

3. method according to claim 1 is characterized in that, confirms that (414) said normalization Δ TOA comprises:

Confirm average corresponding to the said Δ TOA of said a plurality of blades;

Average based on corresponding to the said Δ TOA of said a plurality of blades is confirmed constant; And

Utilize average, said Δ σ _ 1 and the said constant of said Δ TOA, said Δ TOA to confirm said normalization Δ TOA.

4. method according to claim 3 is characterized in that, confirms that (414) said normalization Δ TOA comprises to utilize following equality to confirm said normalization Δ TOA:

Wherein be the normalization Δ TOA corresponding to blade k at time constant t place, be the Δ TOA corresponding to said blade k at said time constant t place, and be Δ σ _ 1 at said time constant t place, and K be a constant.

5. method according to claim 1 is characterized in that, confirms that (418) said correction Δ TOA comprises:

Confirm (422) corresponding in said a plurality of blades each move a side-play amount; And

Deduct from said normalization Δ TOA that (428) are said moves a side-play amount, thereby produce said correction Δ TOA.

6. method according to claim 5 is characterized in that, when said Δ σ _ 2 are to confirm a said side-play amount of moving when move the first time after startup greater than predetermined threshold value or said a plurality of blades (12).

7. method according to claim 5 is characterized in that, confirms that (422) said side-play amount of moving comprises:

Reception is corresponding to each the one or more normalization Δ TOA (602) among one or more in said a plurality of blades;

Utilize said one or more normalization Δ TOA to confirm one or more correction Δ TOA (604);

Confirm first average of (612) said one or more normalization Δ TOA;

Confirm second average of (614) said one or more correction Δ TOA; And

From said first average, deduct (616) said second average, thereby produce a said side-play amount of moving.

8. method according to claim 7 is characterized in that, when said one or more blades are not that said one or more normalization Δ TOA is confirmed the time in operation under transient state.

9. method that is used to monitor the health status of a plurality of blades (12) comprises:

Confirm (106) Δ TOA corresponding to said a plurality of blades (12);

Calculate (404) standard deviation corresponding to the said Δ TOA of said a plurality of blades;

Confirm (412) one or more normalization Δ TOA corresponding to said a plurality of blades;

Confirm the standard deviation of (416) said normalization Δ TOA;

Utilize the said standard deviation of said normalization Δ TOA to confirm (418) Δ σ _ 3 with the standard deviation before the normalization Δ TOA; And

Confirm (428) correction Δ TOA based on said Δ σ _ 3.

10. system comprises: processing subsystem (22), its:

Confirm (106) Δ TOA corresponding to a plurality of blades;

Utilization is confirmed (404) standard deviation corresponding to the said Δ TOA of said a plurality of blades;

Utilize said standard deviation and primary standard deviation to confirm (412) Δ σ _ 1;

Utilize said Δ σ _ 1 to confirm (414) one or more normalization Δ TOA corresponding to said a plurality of blades;

Confirm the standard deviation of (416) said normalization Δ TOA;

Utilize the said standard deviation of said normalization Δ TOA to confirm (418) Δ σ _ 2 with the standard deviation before the normalization Δ TOA; And

Confirm (418) said one or more correction Δ TOA based on said Δ σ _ 2 corresponding to said a plurality of blades.

Images